Site Reliability Engineer (AI)

TL;DR

Site Reliability Engineer (AI): Driving the end-to-end reliability of the Tinker fine-tuning API with an accent on distributed training systems, production observability, and incident response. Focus on hardening multi-tenant isolation, maximizing GPU utilization through resource scheduling, and building resilient recovery systems for long-running distributed jobs.

Location: Based in San Francisco, California. Relocation support provided.

Salary: $350,000 – $475,000 USD

Company

hirify.global is building the future of collaborative general intelligence, creating the Tinker fine-tuning API to empower researchers and developers to customize frontier AI models.

What you will do

• Define and own end-to-end reliability, encompassing CI/CD flows, production observability, and incident response.
• Develop Service Level Objectives (SLOs) for distributed training systems to balance reliability, latency, and velocity.
• Design and implement comprehensive monitoring and observability across the full training path.
• Lead incident response for the Tinker platform, ensuring rapid recovery and implementing systematic improvements.
• Harden multi-tenant isolation and resource scheduling to maximize GPU utilization without compromising data separation.
• Collaborate with security teams to identify and address production vulnerabilities.

Requirements

• Bachelor's degree or equivalent experience in Computer Science, Engineering, or a similar field.
• Professional experience in distributed systems, cloud infrastructure, or site reliability engineering.
• Proficiency in writing software to automate reliability tooling and solve complex infrastructure problems.
• Proven track record in production incident response, postmortems, and reliability improvement.
• Strong communication skills and experience coordinating across engineering and research teams.
• Must be based in or able to relocate to San Francisco, California.

Nice to have

• Deep experience operating production cloud services at scale.
• Background in distributed training frameworks and understanding of infrastructure failure modes in training.
• Experience building checkpoint and recovery systems for long-running distributed workloads.
• Expertise in operating and tuning Kubernetes clusters handling heterogeneous GPU workloads.

Culture & Benefits

• Generous health, dental, and vision benefits.
• Unlimited PTO and paid parental leave.
• Relocation support for candidates moving to San Francisco.
• Visa sponsorship availability for the right fit.